Posted by jskenner on 11/11/2010 9:47:00 AM (view original):
Exactly.  It's as if the engine only looks at what occurred in the game previous to the shot, not considering other factors (such as distribution, defender quality, fatigue, etc), then independently adjusts odds of a make based on if the shooter had made or missed previous shots.  So even though a tired player forcing a shot against a great defender might have a 35% chance of a make, the fact he's 6-19 so far produces a 58% chance of a make.  Not good.

No JSK, I'm not sure we disagree too much at all.
If your above post is what the engine is doing... then I agree as well that it needs a change... maybe only a small one... maybe not.
My point, or at least one of my points was,,, that I'm not sure that how you discribed it above is what's actually happening here. The guy did go 0-4 to start the game... probably against a good FRESH double team of starters from the #1 team in Allen DIII..... and trust me... playing the devil's advocate for HD is not one of my normal job's... I just happen to know something about this conference and the teams/players involved.... and my point was that I'm not sure the numbers everyone's using to point out an engine flaw can't be explained....or at least partially explained.

I hear you, mizzou.  I'm not arguing it from the original single-player instance.  I'm arguing it from my experience since the change.  Before the change, shooting was fairly inconsistent, and now it's much more consistent overall, especially for high-ability scorers, regardless of distribution.  And I THINK ku and and jet are on to something.  But I'm not convinced of either side.

Good news.  This terrible Sim AI team www.whatifsports.com/hd/TeamProfile/Ratings.aspx?tid=7836  ended up beating Harold Baldwin's team in a big upset last night.  And Baldwin was terrible against the terrible Sim AI team:  5 for 20 shooting, 3 for 13 from three point range: www.whatifsports.com/hd/GameResults/BoxScore.aspx?gid=6356581

I dont mind how good this player is, but the lack of variance is a complete joke.  I assume that this itself isn't an outlier sample of low variance (with enough 16 game samples, you're bound to have a sample of extremely low variance even in a high variance environment - but somehow I think this is more indicative of the norm as opposed to an outlier - since this has been an explicit goal by seble). So with the assumption that this is the norm and a desirable one (by seble) at that - Booooooo Seble!  Variance is our friend.  We should attack the fear of variance with education instead of with unnecessary modifications to the sim to align it with misconceptions about what is otherwise standard variance within small sample sizes.  

speaking as a casual user who plays this game on and off and wants a "realistic" experience, this news really turns me off.   this is sort of similar to injuries, which are turned way down in this game, for no apparent reason except to prevent whining.  I feel as if those who prefer the recruiting aspect of the game to the gameplanning  are getting catered to. 

I know when I sign up for this game that I have zero chance of winning a national title without dedicating myself for several seasons if at all, so the chance of gameplanning for an "upset" of someone who has gamed the RPI is what drives me when I play this game.  i don't know why upsets/variance should be turned down.  In real life,  if the 10th best team in the country plays the 200th best team, they will  *only*  have a  ~90-95% chance of winning,  a 12-15 point favorite has only a 86-92% chance.  Does anyone have any estimate how often a "competitive" underdog (say a C/C- prestige team with an active coach) beats an A team?  I suspect that it happens a lot less often that it does in real life.

variance is our friend, its what makes sports fun.  please have VARIANCE.

KUJayhawk is right.  The new engine was supposed to CORRECT the PROBLEM of 1-man shows lighting it up from 3-point range no matter how much distro they are given.  Nobody at any level of basketball shoots 45%+ from 3-point range over the course of a season if they're taking 8-10 three pointers a game.  (Plenty do it that don't shoot very often... similar to the -1 setting).  It just shouldn't happen.  Submit a ticket, KU.